The purpose of this study is to increase knowledge about the psychopathology of depression and anxiety by means of a longitudinal multidimensional assessment of cognitive and neuroendocrine aspects in a medication-free population. The specific aims of the study are to test whether cognitive phenomena and neuroendocrine abnormalities can differentiate individuals with clinical diagnoses of depression and anxiety from each other and from a normal comparison group, to investigate the association of neuro endocrine dysfunctions and cognitive factors in depression and anxiety disorders, to extend the knowledge of the association between suicidal intent and cognitive phenomena, and to determine changes in neuroendocrine function and cognitive factors over the course of cognitive therapy. This study is intended to refine the assessment and understanding of depression and anxiety disorders and to identify differential treatment targets for the two disorders. Eighty outpatients assigned a primary diagnosis of major depression (40 endogenous and 40 nonendogenous) on the basis of a standardized interview will be compared to 40 outpatients assigned a primary diagnosis of anxiety disorder (generalized anxiety disorder or panic disorder) and 40 normal individuals recruited through the media. The assessment battery will include a measure of overal psychopathology, overall measures of depression and anxiety, multi-level cognitive measures (assessing labile "automatic thoughts" and relatively stable underlying assumptions), and three neuroendocrine challenge tests. All participants will be administered the psychological scales at intake and after 6, 12, and 24 weeks. Patients will undergo the neuroendocrine tests at the same points, and normal controls will undergo those tests only at intake and 12 weeks. All patients will receive standard cognitive therapy. Discriminant analyses will be used to test the ability of the cognitive and neuroendocrine measures to discriminate depression and anxiety, and repeated-measures analyses of varience will be used to test for changes in measures of depression, anxiety, cognition, and neuroendocrine function over the course of cognitive therapy. In addition, cognitive and neuro-endocrine characteristics of responders and nonresponders to treatment will be compared.